* Studio: match llama.cpp SWA cache sizing
* Studio: account for batch-capped SWA ubatch
* Studio: match llama.cpp KV stream padding
* Match llama.cpp batch and FA-off cache sizing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Skip unusable compact SWA slot saves
* Align KV planning with launched server
* Match cache type casing and narrow the compact SWA slot-save skip
The launcher tested the requested cache type case-sensitively while the budget
lowercases it via _planned_main_cache_types, so a Q8_0 request emitted no
--cache-type flag and llama.cpp ran f16 while the estimate priced q8_0 (1.01 GiB
under-reserved on a 27B SWA model at ctx 32768 with 4 slots).
The compact SWA slot-save skip keyed on the sliding window alone, but the
estimator's SWA path also requires key/value length. phi3 GGUFs report a window
without those dimensions and llama.cpp runs them non-SWA, so their slots restore
fine and were being skipped.
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* Fix GGUF tool chat server recovery
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cover MTP precedence and loosen the replay assertion for PR #7424
Add a regression test for the MTP branch of the tool-loop respawn retry: the
file-wide _make_backend stub forces _maybe_recover_from_mtp_crash to False, so
nothing exercised the case where an MTP crash reload is already claimed and an
ordinary same-config respawn must not run on top of it. Cover both the next
tool-loop request and the final synthesis pass.
Replace the whole-payload equality assertions with a field-wise check. Comparing
the full dict pins max_tokens to the value derived from the dead server's
effective context, so a later fix that rebuilds server-derived defaults after a
respawn would read as a test failure rather than an improvement.
Document that the one-retry budget is per model request, not per chat turn.
* Recover from prefill-time deaths and stop respawn racing the MTP reload
Two gaps in the tool-loop respawn retry, both reproduced before fixing.
A child that exits during prefill has already accepted the socket, so httpx
raises ReadError, WriteError or RemoteProtocolError rather than ConnectError.
Those all arrive before the response opens, which is exactly the window where a
replay is safe, but the helper only caught ConnectError and gave up. Widen the
catch to NetworkError plus RemoteProtocolError. Timeouts stay excluded on
purpose: they mean the server is slow, not dead, and retrying one would spend
the 20 minute first-token budget twice. Windows resets connections where Linux
refuses them, so this also covers the common Windows presentation.
_maybe_recover_from_mtp_crash returns False both when the crash is not an MTP
crash and when an MTP-free reload is already in flight. Callers read that as
permission to respawn, so _respawn_if_dead replayed the crashing MTP kwargs and,
by replacing the process, made the in-flight reload abort on its own newer-load
check. Skip the respawn while that reload owns the corpse. The guard lives in
_respawn_if_dead so the plain chat path gets it too.
Regression tests for both, including a guard against retrying prefill timeouts.
* Release the MTP single-flight claim when the reload never starts
_mtp_runtime_fallback_in_progress is claimed before the reload thread exists, and
only that thread's finally clears it. Two statements ran in between with no unwind
path: re-reading _last_load_kwargs, which an unload can null underneath us, and
Thread.start(), which raises under the thread exhaustion that is exactly the
pressure killing llama-server in the first place. Nothing else ever resets the
flag, so a failure there latched it for the life of the process.
That was survivable before, since respawn ignored the flag. It is not now: the
guard added in db78184be keys off the flag alone, so a latch would silently
disable auto-respawn for every later model, including plain non-MTP ones. Read
the kwargs and process once before claiming, and release the claim if the thread
cannot start.
Restore the whole-payload equality assertions. Comparing field-wise was meant to
leave room for rebuilding server-derived defaults on replay, but the payload is
built once before the retry and re-sent unchanged, so the looser check only
dropped seven real keys and added a vacuous seed comparison.
Also correct the docstring: llama-server flushes its 200 at slot start, so a
death during decode arrives with the response already open. The pre-header window
this covers is an upload still in flight or a request waiting behind busy slots.
* Confirm the child exited before spending the retry
A closing llama-server can beat its own exit status: the socket error arrives while
poll() still reports the process running. _respawn_if_dead then took the alive
branch, handed back the stale _healthy, and the caller read that as a successful
respawn and spent its single retry on the same corpse. When that retry failed,
attempt was no longer 0, so no respawn ever happened and the turn died, with a log
line claiming a respawn that had not occurred. The window matters most for the
pre-header ReadError and RemoteProtocolError shutdowns the retry now covers.
Wait a bounded second for the exit status before calling the child alive. The same
race is already conceded in _maybe_recover_from_mtp_crash, whose recovery thread
polls for 5s because the error can arrive a beat early; 1s here because this runs
on the request path, and a genuinely live server, including one a concurrent caller
has just respawned, still returns promptly.
* Tighten the recovery comments
* Harden the respawn path around concurrent unloads and replacements
Two problems with the reap grace loop, both found by review.
Skip the grace when the server was already replaced. A caller queued on
_respawn_lock behind someone else's respawn woke holding the healthy replacement,
could not tell it from the child its own request had used, and waited out the full
grace. That sleep is under the lock, so the waits serialised: four concurrent
generations cost roughly three grace periods before any retry began. Capture the
process before taking the lock and return early once it has been swapped.
Do not respawn a server that is being torn down on purpose. unload_model() sets
_cancel_event and only clears _last_load_kwargs after the kill, so a request losing
its connection mid-unload could watch that deliberate exit through the grace loop,
read the stale kwargs and load the model straight back; a model switch landing
during the wait was reverted the same way. Re-check the cancel flag and the process
identity under _serial_load_lock before capturing the replay kwargs, matching what
the MTP-crash reload already does.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Tighten the respawn comments
* Do not charge the reap grace to a server that is still serving
The grace loop added for the not-yet-reaped race waits on poll(), which for a
live child never returns, so every transient transport error paid the full
_RESPAWN_REAP_GRACE_S. That sleep is held under _respawn_lock, so the cost
serialised: measured 1002 ms for one caller and 8.02 s for eight concurrent ones,
against 0 ms on main. A working install pays this, not a broken one.
A llama-server's listening socket dies with the process, so a loopback connect
separates the two cases in microseconds. Probe it first and return immediately
when the port still accepts; fall through to the grace only when the port is
gone, which is the case the grace exists for. Back to 0.7 ms for one caller and
0.00 s for eight.
Cross-checked on real hardware over Qwen3.5-2B, Llama-3.2-1B, Gemma-3-4B with
mmproj and Qwen3-30B-A3B: decode throughput within noise of main (-0.06%, -3.71%,
+2.57%, +0.29%, against a 54-232% spread between rounds of a single run), output
byte-identical on every round, tool-path recovery restored on the three families
whose model calls the tool, and plain-chat recovery still working on all four.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make the respawn lose to a deliberate unload in every window
Two follow-ups on the respawn path, both reproduced first.
Check _cancel_event before the socket fast path. unload_model sets the flag before
it kills, so the child is still accepting when the probe runs; returning the stale
_healthy there aims the retry at a server that is deliberately going away.
Close the unload TOCTOU. The old cancel check sat under _serial_load_lock, which
unload_model never takes, so an unload could land entirely between that check and
load_model and the captured kwargs would restart a model the user had stopped.
Snapshot the kwargs, the flag and a new _unload_epoch together under _lock, the
lock unload does hold, so a teardown is either wholly before the snapshot or
wholly after it. load_model clears _cancel_event on the way in, so the epoch is
the only evidence that survives; when it moves during the reload the replacement
is unloaded again rather than left running.
_lock stays uncontended across load_model, which would deadlock a plain Lock and
block /status for the length of a load. Error-path latency is unchanged: 0.6 ms
for a live server and 0.00 s for eight concurrent callers.
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: danielhanchen <unslothai@gmail.com>
Co-authored-by: danielhanchen <danielhanchen@gmail.com>
* Studio: GPU memory dropdown — llama.cpp --fit on and manual gpu-layers/cpu-moe
* Studio: simplify GPU memory changes (reuse ParamSlider, GPU_LAYERS_ALL, loadedGpuMemoryFields helper)
* Studio: GPU picker — choose which GPUs a GGUF model loads on (gpu_ids)
* Studio: simplify GPU picker (share /api/system fetch, validate gpu_ids)
* Studio: GPU picker review fixes (gate relative indices, no cross-model leak, validate, types)
* Studio: group GPU controls under a collapsible GPU section
* Studio: GPU feature review fixes (fix fit-ctx test, behavior-test the floor, comment accuracy)
* Studio: make GPU a top-level settings section (not nested under Model)
* Studio: flatten GPU controls into the Model section, group by GPU/context/generation
* Studio: move GPU Memory to the bottom of Model with its dependent controls beneath it
* Studio: move GPU Memory below Tensor Parallelism and GPUs below GPU Memory
* Studio: tighten GPU Memory and GPU Layers tooltip copy
* Studio: fix fit-mode context slider track-click, restore GPU Memory tooltip, shorten fit dropdown label
* Studio: GPU Memory tooltip one mode per line, briefer
* Studio: note HIP_VISIBLE_DEVICES (ROCm) in the GPUs picker tooltip
* Studio: narrow the GPU Memory dropdown to fit the shortened label
* Studio: use 'llama.cpp --fit' in the GPU Memory tooltip for consistency
* Studio: allow Tensor Parallelism in Manual GPU mode
* Studio: graduated MoE-on-CPU offload (--n-cpu-moe) replacing the all-or-nothing toggle
* Studio: size the MoE-offload slider for staged (deferred-load) models
* Studio: share one GGUF header walk for the context-length and MoE-count readers
* Studio: size the GPU Layers slider for staged models (one staged-header read)
* Studio: move Tensor Parallelism below the GPUs picker
* Studio: GPU split (--tensor-split) per-GPU model share in Manual mode
* Studio: tolerate whitespace in GPU split input, move it below GPU Layers
* Studio: rename the GPU split control to "Split ratio"
* Studio: Split ratio sends explicit even input; fix blank=free-VRAM (not even) copy
* Studio: tighten llama.cpp --fit VRAM margin with --fit-target 512
* Studio: GPU memory review fixes (rollback re-baseline, single-GPU TP gate, accurate copy)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: move Split ratio below MoE Layers on CPU
* Studio: address PR review (fix GPU-info hydration race, share fit context-length across load paths)
* Studio: address codex review (manual single-GPU TP guard, GPU-aware spec defaults in fit/manual, GGUF-only context/preference)
* Studio: address codex review round 2 (gpu_present seed, single-GPU tensor-split guard, staged manual-knob reset, strip inherited offload flags)
* Studio: address codex review round 3 (strip inherited --n-cpu-moe, CPU-fallback warning in Manual mode)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: address codex review round 4 (preserve pinned fit context across a later Apply)
* Studio: address codex review round 5 (honor GPU picker for diffusion GGUFs, clear fit pin on cross-model switch)
* Studio: preserve the pending GPU Memory mode when staging a model
* Studio: pin diffusion GPU device order and reset GPU-memory state for diffusion loads
* Studio: address codex review round 6 (fit-Auto rollback context, preserve manual non-tensor split modes, persist GPU mode on load not select)
* Studio: persist the applied GPU Memory mode, not the requested one (skip diffusion loads)
* Studio: replace Manual-mode split-ratio field with per-GPU layer sliders
* Studio: clarify per-GPU layer split hint for tensor-parallel mode
* Studio: address codex review round 7 (allow GGUF gpu_ids past the legacy guard, replay GPU-memory fields on respawn)
* Studio: address codex review round 8 (size the validate preflight like the load in fit mode, across both load paths)
* Studio: skip the training-OOM guard for llama.cpp --fit GGUF loads (they spill to RAM)
* Studio: drop the now-redundant compare-path validate sizing (the --fit guard skip makes it moot)
* Studio: address codex review round 9 (keep the training guard for fit loads, forward gpu_ids to validate, strip inherited manual tensor-split)
* Studio: address codex review round 10 (gate GPU-memory adoption on is_gguf, record manual knobs only in Manual mode)
* Studio: handle diffusion GGUFs symmetrically in the GPU Memory controls (preserve the standing mode preference, hide the inapplicable mode/TP controls)
* Studio: remember the GPU Memory settings per model
* Studio: consolidate --fit mode and Manual mode into a single Manual mode
* Studio: preserve the per-GPU layer split across GPU Layers changes
* Studio: trim overly long GPU Memory comments
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* address GPU memory config review comments
* trim redundant GPU memory tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Reconcile manual-mode TP drops with the #6659 drop-site invariants
* Preserve quantized KV in manual --fit, charge GGUF companions in full, reconcile GPU pick on load
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Clear stale GPU baseline on non-GGUF loads so it can't read as dirty
* Fix no-context-shift test for the conditional -c flag
* Credit manual GPU-layer offload for cached HF GGUFs
* Reset per-model load knobs on GGUF quant switch
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Strip inherited tensor-split when manual ratio is cleared
* Match auto-load validation to safetensors placement
* Reset editable manual knobs after Auto GGUF loads
* Record a single device for diffusion GPU picks
* Reset per-model GPU knobs before applying saved settings
* Address review comments
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Guard manual tensor splits and keep remembered context on auto-load
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Snapshot compare knobs, seed splits from free VRAM, flag zero-offload loads
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Exempt CPU-only loads from the guard floor and harden compare and reseed paths
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Reach full offload from the layers slider and charge extras drafters in the guard
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Warm the GPU device cache before pick reconciles and disable staged GPU controls
* Align the training guard with inherited extras, spec mode, and compare targets
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Hide GPUs from companion-less zero-offload loads
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Size diffusion picks per device, own manual offload flags, reject XPU picks
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Drop tensor flags at zero layers and exempt CPU-pinned drafters
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Allowlist the zero-layer tensor parallel drop site
* Keep validate and load guards on the same extras and refresh stale baselines
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Drop mismatched manual tensor splits before launch
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Gate XPU picks on the real backend field and harden split and hydration paths
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Weight full GPUs as zero, clamp split shares, and refine the zero-layer mask gate
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Carry fit context across mode changes and align drafter and picker gates
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Catch variant switches, uncached diffusion repos, and text-only mmproj skips
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Check companions on the first device and size native and remote zero-layer loads
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Replace the training guard's precise VRAM modeling with a conservative bound
* Baseline context pins on non-GGUF hydration and reprobe list-seeded staged GGUFs
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Size manual splits by their largest share and preserve resolved context from Default
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Default-deny unsized required companions and price KV at the effective cache dtype
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Reserve MTP draft KV and MLA target-copy in the training guard
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Size tensor-parallel loads per device and show GPU controls for native GGUFs
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Reserve MTP overhead for uncached remote GGUFs and the mmproj runtime factor
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Drop the training-coexistence VRAM estimation this PR added
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Gate remembered load settings to GGUF picks
* Lock the remaining load-time controls during a staged load
* Clear the stale native-path token on compare loads
* Drop a stale guard reference from the zero-offload masking comment
* Seed GPU baselines from the rollback response and drop never-emitted offload flags
* Match validate's training guard to load and keep the native reload token
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Trim verbose GPU-memory comments
* Thread the variants header walk off the event loop, honor device pins on zero-offload, and hold staged GPU edits
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Honor manual placement and classify pinned zero-offload loads
* Close diffusion admission and status hydration gaps
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Check the actual diffusion GPU during training
* Align staged baselines and manual reload dedupe
* Fix GGUF placement and rollback state
* Harden manual GGUF placement boundaries
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Remove unused resolve_tensor_parallel import in llama_cpp.py
The name is used only in llama_server_args.py, routes/inference.py, and tests,
not in llama_cpp.py; the unused hoisted import trips the import-hoist verifier
in the source-lint CI job.
* Fix diffusion GPU dedup and training guard for non-numeric device tokens
The diffusion runner drives only its single lowest device and the backend
records that one device (self._gpu_ids = [sorted(gpu_ids)[0]]), but the reload
dedupe compared it against the full requested list, so a multi-GPU pick that
resolves to the same device forced a needless reload. Normalize the request the
same way for a loaded diffusion model in both _already_in_target_state and the
route _request_matches_loaded_settings.
The chat-during-training coexistence guard called int() on the single-device
token and hard-rejected when it could not parse. A non-numeric token (a CUDA
UUID / MIG handle) now sizes against the whole visible pool like the GGUF guard
instead of falsely blocking the load, and an empty token (a CPU-only runner such
as a CPU diffusion GGUF) is allowed outright since it uses no GPU VRAM.
* Tighten comments added by the GPU memory config changes
* Harden GGUF placement from independent review: VRAM sizing, diffusion TP reset, tensor_split validation
- Training coexistence guard: a single-device runner pinned through an
unresolvable UUID/MIG token was sized against the aggregate visible-VRAM pool,
so a load could pass on capacity it cannot use and then OOM active training.
Size against the worst-case visible device (min free) instead, keeping the
guard's documented default-deny contract. The empty-token (CPU-only runner)
allow path is unchanged.
- Diffusion startup: _start_diffusion_server now resets self._tensor_parallel to
False alongside the other placement resets. A prior tensor-parallel chat load
(process killed but not fully unload-reset) otherwise left /status misreporting
tensor parallelism and made an identical diffusion re-Apply reload against the
stale state.
- tensor_split: reject negative / non-finite / all-zero splits up front. They
were dropped at launch but still compared raw in the reload dedupe, so an
identical Apply reloaded indefinitely.
- Tests: the shared httpx stub was incomplete and, installed via setdefault
before real httpx loaded, broke a combined pytest run (collection errors on
httpx.Response). Import the real installed httpx instead.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: danielhanchen <unslothshared@gmail.com>
Co-authored-by: danielhanchen <danielhanchen@gmail.com>
* Replace standalone Studio wording with Unsloth
Replace the single word Studio with Unsloth wherever it is used as
shorthand for Unsloth Studio in docs, CLI output, UI strings, i18n
locales, workflow display names, comments and docstrings.
Kept unchanged: the full name Unsloth Studio, third party product
names (LM Studio, Visual Studio, Mac Studio), feature names
(Recipe Studio, Fine-tuning Studio and its translations), and all
identifiers such as env vars, commands, paths and filenames.
* Address review feedback on the Studio wording rename
Use "an" before Unsloth where the rename left the article as "a".
Restore the split brand where Unsloth and Studio render as two halves
of the full product name: the onboarding sidebar subtitle and the
IPv6 localhost warning. Scope two messages to the full name Unsloth
Studio where plain Unsloth was misleading: the AMD README bullet and
the CLI studio setup error.
---------
Co-authored-by: Lee Jackson <130007945+Imagineer99@users.noreply.github.com>
Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>
* fix(studio/llama_cpp): disable trust_env on the loopback health probe
_wait_for_health() polls http://127.0.0.1:<port>/health with the default
httpx trust_env=True, so an ambient HTTP(S)_PROXY in the environment is
applied to the loopback request. A proxy that returns 503 for 127.0.0.1
makes every probe fail, so the loop runs until timeout and Studio load
hangs (trust_env=False returns 200 immediately).
Pass trust_env=False so the local readiness probe never goes through a
proxy. This mirrors the existing trust_env=False handling in the sibling
llama_http / external_provider HTTP clients.
* test(offline_gguf_cache): accept trust_env kwarg in fake_get mock
_wait_for_health now calls httpx.get(..., trust_env=False); update the retry test's fake_get to accept the kwarg so it doesn't raise TypeError.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix(studio/llama_cpp): bypass proxies for loopback clients
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix(studio/routes): bypass proxies for llama streams
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: wasimysaid <wasimysdev@gmail.com>
test_safetensors_capability_advertise: detect_reasoning_flags now returns a
reasoning_effort_levels key, so the none-template expectation must include it.
test_tensor_parallel::test_runtime_recovery_reloads_without_mtp: the assertion
raced the recovery thread, which sets _spec_fallback_reason just before its
finally clears _mtp_runtime_fallback_in_progress. Wait for the flag to clear
before asserting.
Co-authored-by: danielhanchen <michaelhan2050@gmail.com>
* Studio: fall back to text-only when a vision projector hard-crashes llama-server
The text-only mmproj fallback (#6075) only fired when llama-server printed a
recognizable projector-format error ("Unknown projector type", exit -6). An
installed llama.cpp that predates a model's projector can instead SIGSEGV
(exit -11) with no parseable output, e.g. unsloth/Qwen3.5-4B-MTP-GGUF +
mmproj-F16 on an older gfx1151 prebuilt: llama-server crashes on load, the
--fit off retry crashes the same way, and load_model gives up with a hard 500
instead of dropping vision.
Generalize the decision: a vision (--mmproj) launch killed by a signal (POSIX
returncode < 0, e.g. -11 SIGSEGV / -6 SIGABRT; Windows 0xC0000000+ access
violation) is treated like a projector incompatibility, so the load retries
once text-only. The retry is skipped if a cancel/unload is pending, mirroring
the MTP guard. Clean non-zero exits (bad GGUF, port bind) and hung processes
keep their own handling; non-vision launches are unaffected.
Reproduced and verified on gfx1151 (Radeon 8060S, ROCm 7.2.1): a current
prebuilt (llama.cpp b9596) loads the exact model + args fine, confirming the
crash is a stale prebuilt. With a wrapper that SIGSEGVs on --mmproj, Studio now
recovers: the load returns 200 (is_vision=false) and serves at ~31 tok/s
text-only instead of failing. New _is_signal_crash helper plus tests pin the
decision.
Also normalize a few em-dashes to ASCII punctuation in existing comments.
* Studio: refine mmproj hard-crash fallback (signal scope + last argv)
- Limit _is_signal_crash to genuine program faults (SIGSEGV, SIGABRT,
SIGILL, SIGFPE, SIGBUS) and Windows 0xC0000000+ statuses. SIGKILL,
SIGTERM and SIGINT no longer count, so an OOM-killer, unload or
supervisor kill is not masked as a projector incompatibility.
- Strip --mmproj from the last attempted argv so the text-only retry
keeps --fit off / --spec-default instead of resurrecting the original
spec flags (matters for MTP vision models on an older llama.cpp).
- Drop stray temp files committed by mistake and gitignore the "~" dir
so they cannot be re-added.
* Studio: tighten comments in mmproj hard-crash fallback
* Studio: retry --flash-attn off before dropping vision on a startup crash
When llama-server hard-crashes at startup, the recovery chain now tries the
least-destructive mitigation first. Flash-attention kernels SIGSEGV at load on
some ROCm/GPU builds (often inside the vision tower's attention); disabling
flash attention keeps BOTH vision and MTP, so a hard program fault with
--flash-attn on now retries once with --flash-attn off before the MTP-drop or
the text-only (mmproj-strip) fallbacks. _is_signal_crash already gates this to
genuine faults (SIGSEGV/SIGABRT/SIGILL/SIGFPE/SIGBUS), so an OOM-kill or unload
(SIGKILL/SIGTERM/SIGINT) does not trigger a retry.
Field context: a gfx1151 user crashes loading a vision GGUF even on the latest
prebuilt, so an update cannot help, and the same model and args load fine on
another gfx1151 box, pointing at a runtime/flash-attn fault. New
_with_flash_attn_off helper plus tests. Verified on hardware with a wrapper
that SIGSEGVs on --flash-attn on: Studio recovers with is_vision=true (vision
and MTP intact) instead of failing or losing vision.
* Studio: name the OOM kill on a too-large model load
When the OS kills llama-server with no diagnostic output (SIGKILL/SIGTERM,
almost always the OOM killer, e.g. a BF16 model too large for the WSL VM's
RAM cap), the recovery ladder correctly does not retry an external kill, so
this is the message the user sees. It fell through to the generic "is the
GGUF valid / out of memory" text. Make it actionable: name the signal and
point at a smaller or more quantized GGUF, a lower context length, or raising
the WSL memory limit. Output-based diagnoses still win and a hard fault keeps
the generic fallback.
* Studio: refuse a model too large for system RAM on a unified-memory APU
On gfx1150/gfx1151 APUs the weights load into shared system RAM (GGML
unified memory). _get_gpu_free_memory reports the full ROCm/APU budget as
free (often ~100 GB), but under WSL the VM's RAM cap is the real ceiling.
Studio trusted the budget, spawned a load larger than RAM, and the OS killed
it mid-flight, taking the Studio process with it (a silent "Terminated" with
no error, the model resident in RAM not VRAM).
Add a pre-flight guard on the APU path: if the weights exceed available
system RAM (psutil, then /proc/meminfo), refuse before spawning with a clear
message (smaller/more-quantized GGUF, lower context, or raise the WSL memory
limit). Weights only so KV/context auto-reduction is not double-counted;
unknown RAM never refuses; non-APU and discrete-GPU paths are untouched.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: refine recovery ladder (keep diagnosed errors, flip all flash-attn)
Two review points on the hard-crash recovery ladder:
1. The signal-only text-only fallback stripped --mmproj on any hard fault,
even when llama-server had already printed a non-projector cause (an OOM
such as "cudaMalloc failed: out of memory", an unsupported architecture, or
a tensor-parallel limit). That masked the real error and told the user to
update llama.cpp for vision. New _output_has_nonprojector_diagnostic gates
the signal path: it fires only when no such marker is present, so a bare
SIGSEGV with no output still retries text-only, but a diagnosed OOM surfaces
the real error instead of silently dropping vision.
2. _with_flash_attn_off only flipped the first --flash-attn. llama.cpp is
last-wins, so a leftover enable from extra_args (--flash-attn on, -fa on, or
the = form) could keep flash attention on and re-crash the retry. It now
flips every occurrence and returns None only when nothing is flippable.
test_llama_cpp_mmproj_fallback.py and the classification/APU suites: 103 passed.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: pass the text-only retry's exit code to the failure classifier
When the text-only fallback retry itself fails, read its exit code before
_kill_process() clears it and forward it to _classify_llama_start_failure, so an
OS-killed retry surfaces the actionable out-of-memory message instead of the
generic one (matching the primary failure path).
* Studio: scope APU RAM guard to selected GPUs, count MTP drafter, neutral SIGTERM
Three refinements to the startup recovery work in this PR:
- The unified-memory APU RAM guard fired whenever any visible GPU was a
gfx1150/gfx1151 APU, so on a mixed APU+dGPU host it could refuse a valid
load placed on the discrete GPU. Scope _amd_apu_wants_unified_memory to the
selected gpu_indices (physical ids, mapped via CUDA_VISIBLE_DEVICES like
_is_datacenter_gpu); None still means every visible GPU. Applied to both the
RAM guard and the GGML_CUDA_ENABLE_UNIFIED_MEMORY env set.
- The RAM guard counted only the main GGUF plus mmproj, so a separate MTP
drafter (also resident in unified system RAM, even when offloaded to CPU)
could push the load past the RAM cap and still get OS-killed mid-load. Add
the drafter weights to the APU RAM total.
- The startup classifier reported SIGTERM (-15) as 'most likely out of memory',
but SIGTERM is also how an unload/cancel or a supervisor stops the server.
Keep the OOM wording for SIGKILL (-9, the OOM killer) and report -15
neutrally.
Tests updated/added accordingly.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: address review on the APU guard and decode-probe ladder
- Map APU physical ids via the active ROCm mask (HIP, then ROCR, then CUDA),
mirroring _get_gpu_memory, so a HIP_VISIBLE_DEVICES-selected APU is matched.
- Only add the MTP drafter to the APU RAM total when MTP will actually engage,
so a stale LLAMA_ARG_SPEC_DRAFT_MODEL cannot refuse a non-MTP load.
- After an MTP first-decode hard fault, retry --flash-attn off (keeps MTP)
before dropping speculative decoding, matching the startup rung.
- Fold the --flash-attn= / -fa= rewrite into one branch.
Tests: tensor-parallel decode-probe assertion updated for the FA-off rung.
* Studio: tighten two comments in the APU guard and RAM preflight
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: refine flash-attn retry and APU RAM guard per review
- _with_flash_attn_off now decides on the effective last-wins value: it returns
None when FA is already off (no wasted retry), and neutralizes a bare
--flash-attn / -fa (which llama.cpp reads as on) so the retry cannot re-enable
it. Length is preserved so downstream index slices stay valid.
- _amd_apu_wants_unified_memory uses 'gpu_indices is not None' so an empty
selection is respected (not treated as all-visible).
- The APU RAM refusal now checks the base model only (main + mmproj); an
optional MTP drafter is dropped by the existing MTP-drop fallback rather than
causing a hard pre-spawn refusal of an otherwise loadable model.
Tests: bare-flag / effective-off / empty-selection / HIP-mask cases added.
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Disable MTP speculative decoding under tensor parallelism
Follow-up to #6040 (Studio tensor-parallel support).
MTP-draft speculative decoding plus --split-mode tensor crashes the CUDA
flash-attn kernel at decode time. The startup /health probe only checks that
llama-server comes up, so the existing MTP-drop fallback (keyed on startup
health) never fires and the server dies on the first generation instead.
Gate MTP off when a tensor attempt actually engages: this runs before the
VRAM planner (so no drafter memory is reserved) and before the speculative
flag build (so no --model-draft / --spec-type is emitted). Ngram modes use no
draft model and are kept, and mtp+ngram degrades to ngram rather than off. The
layer-split fallback re-runs with tensor_parallel False and restores MTP.
The reason is surfaced as spec_fallback_reason "tensor_parallel" so the
settings sheet explains why MTP is off instead of prompting a llama.cpp update.
Verified on unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q4_K_XL across 4x B200: the
load now emits --split-mode tensor with no MTP flags and generation completes
without the prior decode crash.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Make tensor-parallel MTP gate test format-independent
The assertion pinned the multi-line `speculative_type = (` form, but ruff
collapses it onto one line, so match `speculative_type =` instead.
* Recover from MTP+tensor-parallel crashes at runtime instead of banning MTP
MTP-draft speculative decoding under --split-mode tensor usually works, but can
crash llama-server's CUDA flash-attn kernel at decode time (the prompt-cache
checkpoint-restore path). The earlier fix statically disabled MTP whenever
tensor parallelism was on, which is not future-proof and gives up the MTP
speedup even though it normally works.
Replace the static ban with a try/recover, mirroring the existing load-time
MTP-drop fallback:
- Load-time decode probe: after the server passes /health under tensor +
MTP, run one tiny /completion to exercise the draft path. A failure flips
the load unhealthy so the existing fallback respawns with --spec-default.
Catches a hard incompatibility that crashes on the first decode.
- Generation-time recovery: snapshot the load kwargs after a healthy load,
and if llama-server exits mid-generation while MTP + tensor parallelism were
active, quietly reload the same model with speculative decoding off (one
single-flight background reload) and surface spec_fallback_reason=runtime_error.
Catches the rare mid-generation crash the probe and load-time fallback miss.
No persistent ban: a later fresh load re-tries MTP, so this self-heals if a
future llama.cpp supports the combo. Verified on gemma-4-26B-A4B + 4x B200:
MTP runs normally, and killing llama-server mid-generation reloads it without
MTP and serves the next request cleanly.
* Address review feedback on the MTP runtime fallback
- Authenticate the decode probe: direct-stream mode runs llama-server with
--api-key, so the unauthenticated /completion probe got a 401 and falsely
dropped MTP. Attach the same bearer auth the other internal requests use.
- Re-check the cancel flag inside the recovery thread after the death poll,
so an /unload that races the reload can't resurrect the dropped model.
- Schedule the no-MTP recovery on the connection-error paths it was missing:
generate_chat_completion's ConnectError branch, the OpenAI passthrough
typed (RemoteProtocolError/ReadError/CloseError) stream catch, and the
Anthropic passthrough generic stream catch. Previously a server that died
before reconnect, or a typed mid-stream error, skipped the reload.
* Cover every request path with the MTP+tensor crash recovery via a watchdog
The runtime MTP-crash recovery only fired from request handlers that
observed the failure, so the direct llama-server proxy endpoints
(/v1/completions, /v1/responses, the OpenAI/Anthropic passthrough
transports) -- and a crash with no request in flight -- could leave a
dead server. Add a single background watchdog, armed only on a healthy
MTP + tensor-parallel load, that polls the subprocess and routes an
unexpected death into the existing single-flight no-MTP reload. It is
stopped inside _kill_process (the one deliberate-termination chokepoint)
so a planned reload/unload is never mistaken for a crash, and re-checks
the stop flag after a detected exit to close the kill-vs-poll race. The
reload turns MTP off, so the replacement server arms no watchdog and the
fallback cannot loop; a later fresh load still re-tries MTP.
* Harden MTP+tensor crash recovery: stale-load race, pass-through MTP, requested mode
Address review findings on the runtime MTP-crash recovery:
- Stale-load race: the recovery thread snapshotted the crashed load, waited up
to 5s for the process to confirm dead, then only checked the cancel flag
before replaying load_model. A concurrent user load clears that flag, so the
stale snapshot could reload the old model over the user's new one. Make the
load lock re-entrant and run the staleness check (cancel + same process +
unchanged snapshot) under it, atomically with the reload.
- Pass-through MTP: MTP can also be requested via a user --spec-type in
extra_args or LLAMA_ARG_SPEC_TYPE, where Studio emits no spec flags and
_speculative_type stays unset, so the probe/watchdog/recovery never engaged.
Track _mtp_runtime_fallback_active from the actual launched config and gate on
it; on the no-MTP reload, append a last-wins --spec-default so the replay drops
MTP regardless of source (and the load-time fallback does the same).
- Requested mode: the off-reload reset _requested_spec_mode to off, so after a
status refresh the UI showed a bare Off with the runtime-error note suppressed
and would not retry MTP. Restore the original requested mode after the reload,
matching the startup MTP fallback.
- Snapshot the extra_args list by value so a caller mutating it cannot corrupt
the recovery snapshot.
Tests: test_tensor_parallel.py + test_llama_server_args.py green (303 passed).
* Trim verbose comments in the MTP+tensor crash recovery
Tighten the docstrings and inline comments added for the runtime MTP recovery
(watchdog, probe, reload, gating) to succinct one/two-line forms; no code
change (verified comment-only).
---------
Co-authored-by: danielhanchen <michaelhan2050@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* studio: reserve MTP draft VRAM in GGUF auto-fit
Auto-fit advertised a context (for example ~110k for the Qwen3.6-27B MTP
GGUF) that fit on paper but OOMed mid-generation or during tool calls once
MTP speculative decoding was active. The MTP draft path's VRAM was reserved
as a flat 5% of total VRAM, which tracks neither of the two real costs: the
MTP head keeps its own attention KV cache that grows with context, and the
speculative verification buffer grows with --spec-draft-n-max. On the
hybrid Mamba/attention Qwen3.6 models the main KV is small, so auto-fit
happily kept a near-native context while the draft path pushed the load
over budget at runtime.
Replace the flat fraction with a byte-accurate, context- and n_max-aware
reserve sized from GGUF dims: draft KV from nextn_predict_layers and the
attention dims at f16 (llama.cpp's MTP draft context uses f16 KV regardless
of the main cache type), plus a verify buffer per embedding-unit per draft
token. The reserve is evaluated per candidate context inside the fit binary
search and added to every pin/fit check, including the tensor-parallel
planner and its even-split decision. Coefficients were calibrated against
llama-server VRAM measurements on the Qwen3.6-27B MTP GGUF (RMS 14 MiB).
The flat fraction remains as a fallback when GGUF dims are unavailable, so
non-MTP loads are unchanged. The budget now also engages when the user wires
MTP through extra args (--spec-type draft-mtp, including chains), reads the
effective draft depth from --spec-draft-n-max or the legacy --draft-max with
extras taking precedence over the first-class field, reserves a separate
drafter's weights when supplied via --model-draft/--spec-draft-model/-md,
and mirrors _build_speculative_flags so it never reserves for MTP the launch
resolver will not emit (needs a head/drafter and a binary that supports
--spec-type mtp).
Adds tests/test_mtp_vram_budget.py.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: total-based VRAM budget + deterministic compute-graph buffer
Build on the byte-accurate MTP reserve with three changes that make the
GGUF auto-fit budget deterministic across architectures and recover usable
context, especially for MTP models on a single tight card.
1. Total-based budget. Cap GPU occupancy at a fraction of TOTAL VRAM rather
than a fraction of FREE VRAM, and raise the fraction from 0.90 to 0.95:
budget = free - (1 - 0.95) * total (per GPU, summed for a pool)
The reserve is now absolute (a fixed slice of the card) instead of
shrinking as the GPU fills, so a partly-used GPU keeps a constant cushion
for compute/CUDA/verify buffers instead of over-promising context and
spilling to CPU at runtime. _get_gpu_memory() reads memory.total alongside
memory.free; _fit_context_to_vram, _select_gpus and the load_model pool
loops thread the totals through. Multi-GPU layer-split pools
sum(free_i - 0.05*total_i); tensor mode reserves per device.
2. Deterministic compute-graph buffer. Replace the flat 5 GB/device tensor
reserve (a magic constant that over-reserved about 8x on a 27B model) with
_estimate_compute_buffer_bytes, sized from GGUF dims and the launch flags:
out = n_vocab * n_ubatch * 4 # vocab-width output buffer
act = 4 * n_embd * n_ubatch * 4 # activation scratch
pipeline_per_device = act + out * (n_parallel - 1)
tensor_per_device = 2*act + out * n_parallel
The buffer is context-independent and scales with --parallel (serving
slots), not with how the model is split across GPUs. It is now reserved in
BOTH multi-GPU paths (layer split folds one buffer into the pooled
footprint; tensor mode reserves it per device). The flat 5 GB stays only as
a fallback when vocab/embedding dims are unavailable. Calibrated against
llama-server measurements (parallel 1/2/4/8 give 36/492/1388/3220 MiB on a
single GPU; about 600 MiB/device tensor); the estimate is a small upper
bound.
3. GGUF parsing. Read vocab size (tokenizer tokens array length) and
feed_forward_length for the compute-buffer estimate.
Effect on the Qwen3.6-27B MTP Q6_K case (MTP on): a single 32 GB card at
about 31 GB free advertises f16 23k to 64k, q8_0 44k to 115k, q4_0 82k to
200k; 2x 24 GB tensor mode recovers the full 262k window for f16 (was about
134k). Validated on hardware: 1x 32 GB f16 at 64768 loads at 29.3 GB / 120
t/s; 2x 23 GB tensor f16 at 262144 loads at 22.2 GB/device / 98 t/s; both
within 0.4% of the estimate. Adds test_compute_buffer.py and updates the
KV/context-fit/MTP-budget tests for the 0.95 constant and the new budget.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: tighten comments in the VRAM auto-fit changes
Condense the docstrings and inline comments added by this PR (internal backend
helpers): drop restated-signature docstrings, fold multi-line block comments to
one or two lines, and remove notes that just repeat the code. No behavior change
(AST-verified comment/docstring-only via comment_tools.py); the backend test
suite is unchanged and green.
* studio: address review findings in the VRAM auto-fit budget
Five fixes from a parallel-reviewer pass on this PR; all confirmed against the
real functions and covered by new tests.
- Tensor mode now honors the total-based VRAM cap. _plan_tensor_parallel took
total_by_idx and budgets each GPU at free - (1-frac)*total, mirroring the
layer-split paths; previously it fit against raw free and could spend the 5%
safety cushion on a partly-used multi-GPU box (reproduced ~3.3 GB over).
- Draft K and V cache types are parsed and accounted independently. A one-sided
override (e.g. --cache-type-k-draft q4_0, V left f16) no longer applies the
small quant to both axes and under-reserves the f16 axis. The embedded-head
formula sizes per axis; the separate-drafter path uses the heavier type so it
never under-reserves.
- The compute-graph buffer honors a user --ubatch / --ubatch-size / -ub override
(parsed and threaded into every _estimate_compute_buffer_bytes call and the
tensor planner); it previously always assumed the 512 default, under-reserving
up to ~8x at --ubatch 4096.
- GPU ranking uses the usable budget (free - (1-frac)*total) instead of raw free
in _select_gpus and both auto-context subset loops, so a more-used large card
no longer outranks a less-used small card that has more usable room.
Adds regression tests for each (tensor total cap, ubatch reserve scaling, split
K/V no-under-reserve, --ubatch parser, usable-ranking GPU selection). Full
targeted backend suite green (321 passed).
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: gate tensor-parallel admission on usable VRAM budget
The tensor-parallel GPU admission filters still used raw free VRAM after
the total-based budget landed, an asymmetric fix: a partly-used large card
can clear the per-device compute-buffer reserve on raw free while its usable
budget (free - (1-frac)*total) does not, so the planner admitted it and the
even split could emit a near-zero weight slice for a GPU that should have
been excluded.
- _plan_tensor_parallel: admit GPUs by usable budget, not raw free (move the
_usable helper above the filter).
- load_model: admit the tensor set by _gpu_usable, and downgrade to layer
split when the pooled usable budget cannot hold weights plus per-device
compute buffers (the planner can only floor the context, not stop an
overcommitted launch).
Adds regression tests: planner drops a GPU whose usable budget is below the
reserve, and a source-level check that load_model admits on the usable
budget and carries the pooled-weight downgrade.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: size the MTP reserve for the user's overriding drafter
A user --model-draft passed in extra_args is appended last and wins at the
llama-server launch, but the VRAM budget preferred Studio's auto-detected
drafter (mtp_draft_path or extras), so a larger custom drafter was
under-reserved. Flip the precedence to extras-first, matching the draft-depth
(n_max) resolution two lines above. Adds a source-level regression test.
* studio: account for MTP reserve in tensor gate, restore 2-col GPU probe
Two issues found by re-review of the prior fix:
- The tensor-parallel capacity gate only checked the model weights against the
pooled budget, not the MTP reserve. A separate-drafter MTP load whose weights
fit but weights + drafter do not could still launch overcommitted in tensor
mode. Add the non-shrinkable MTP reserve (drafter weights + floor draft KV, or
the flat 2 GiB fallback when dims are unavailable) to the gate.
- The nvidia-smi probe was switched to a three-column query (index,free,total)
for the total-based budget but required exactly three columns, so a driver or
mock returning the legacy two-column "index,free" was dropped and the probe
fell through to the real GPUs. Accept two columns (total 0) and treat an
unknown total as the legacy free*fraction in _select_gpus.
Tests: tensor gate asserts the MTP term is included; _get_gpu_memory parses both
two- and three-column output; the existing two-column GPU-detection mocks pass
again.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: keep the VRAM cushion in tensor planning when GPU totals are unknown
_plan_tensor_parallel fell back to raw free VRAM when a GPU's total was
unavailable (a two-column nvidia-smi probe reporting total 0), while
_select_gpus and the load_model ranking both fall back to free*fraction. That
let tensor planning spend the 5% cushion the rest of the fit preserves and
over-advertise context in exactly that path. Align the fallback to
free*_CTX_FIT_VRAM_FRACTION. Updates the no-totals planner test expectations
(now free*frac) and adds a regression test that total 0 keeps the cushion.
* studio: honor LLAMA_ARG_* env overrides and HF draft flags in the VRAM budget
The budget parsed llama-server flags only from the request's extra_args, but the
child process inherits Studio's full environment (child_env_without_native_path_secret
copies os.environ), and llama-server honors LLAMA_ARG_* env vars for the same
options. So a service-level override the child acts on was invisible to the fit,
which could then advertise a context/GPU set that OOMs at load.
- _extra_args_n_ubatch: fall back to LLAMA_ARG_UBATCH (drives the compute buffer;
an unseen 4096 vs the 512 default under-reserves ~8x).
- _extra_args_mtp_draft_path: also recognize the HF draft-repo flags
(--spec-draft-hf/-hfd/-hfrd/--hf-repo-draft) and fall back to
LLAMA_ARG_SPEC_DRAFT_MODEL / LLAMA_ARG_SPEC_DRAFT_HF_REPO. An HF repo isn't a
local file so it can't be sized, but recognizing it routes to the flat reserve
instead of mis-sizing Studio's auto/embedded drafter.
- _extra_args_draft_cache_types: fall back to
LLAMA_ARG_SPEC_DRAFT_CACHE_TYPE_K/_V per axis.
CLI extra_args win over env (they are appended last at launch). Each parser takes
an injectable env for deterministic tests. Adds env-fallback and HF-flag tests.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: review polish - drop non-flag --ubatch, harden GPU probe, document buffer
Non-blocking items from a second review pass; no behavior change in the common path:
- _extra_args_n_ubatch: drop --ubatch; the binary only accepts --ubatch-size/-ub,
so parsing --ubatch implied support it does not have (it would over-reserve for a
launch that fails on the unknown flag).
- _get_gpu_memory: skip a malformed nvidia-smi line instead of letting one bad line
raise and drop the whole NVIDIA probe to the torch fallback.
- _estimate_compute_buffer_bytes: document that the per-slot output-buffer model
assumes a small n_outputs_max (chat decode); it would under-count for
embeddings / --logits-all / reranking, which Studio does not run on this path.
* studio: honor LLAMA_ARG_SPEC_TYPE when deciding the MTP reserve
_extra_args_requests_mtp only checked extra_args, but the child inherits Studio's
env and llama-server honors LLAMA_ARG_SPEC_TYPE. So a service-level
LLAMA_ARG_SPEC_TYPE=draft-mtp would run MTP while the fit skipped the draft
reserve and could advertise a context/GPU set that OOMs at load. Recognize the
env value (CLI still wins). Completes the env-override coverage alongside ubatch,
draft model, and draft cache types. Adds an env regression test.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: reserve VRAM for non-MTP model-based draft modes too
The draft reserve only engaged for MTP. A user passing a non-MTP model-based
draft mode (--spec-type draft-simple / draft-eagle3) with a --model-draft loads
a separate draft model whose weights + KV consume GPU memory, but the fit
reserved nothing and could OOM at load. Engage the existing drafter reserve for
those modes when extras (or LLAMA_ARG_SPEC_TYPE) name a drafter; ngram-* load no
model and are unaffected. Purely additive (reserves where there was none).
Adds parser + gate tests.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: floor quantized embedded MTP draft KV at f16; fix two test issues
Address PR review feedback (three findings):
1. Quantized embedded MTP draft KV was underpriced. The embedded head is a
single draft layer, so llama.cpp cannot amortize quantized-KV overhead over
many layers the way the main model does: a quantized draft KV (e.g.
--spec-draft-type-k q4_0) actually fits LESS context than f16, not more
(ggml-org/llama.cpp#24102, where a collaborator recommends f16 for the draft
KV). Pricing q4_0 at 0.5625 of an element (~28% of f16) under-reserved, so a
quantized override could advertise a context that shrinks or OOMs at load.
Floor the embedded draft KV bytes-per-element at f16 (quantized types priced
as f16, f32 still its full 4 bytes). The separate multi-layer drafter, where
quantization does amortize, keeps the user's real type.
2. test_load_model_reserves_for_non_mtp_draft_modes asserted an exact one-line
source substring that pre-commit black wrapped across lines, breaking CI.
Strip whitespace before matching so the check survives any line-wrapping.
3. test_compute_buffer.py installed a partial httpx stub via setdefault that, if
collected before test_kv_cache_estimation.py, leaked into sys.modules without
HTTPError/Response and could break the transformers introspection tier by
collection order. Adopt the sister file's pattern: only stub when real httpx
is absent, and include the full symbol set.
Updates the affected draft-KV tests to assert the f16 floor.
* studio: guard httpx stub in test_mtp_vram_budget too
test_mtp_vram_budget.py installed a partial httpx stub via setdefault that, like
test_compute_buffer.py before it, lacked HTTPError/Response and could leak into
sys.modules ahead of tests that need huggingface_hub/transformers, breaking the
introspection tier by collection order. Apply the same guard used by
test_kv_cache_estimation.py: only stub when real httpx is absent, with the full
symbol set.
* studio: per-device layer-split reserve, effective spec-type, drafter weights, KV restore
Address PR review feedback (four findings in the auto-fit budget):
A. Reserve the per-device layer-split overhead. A layer (pipeline) split allocates
a fixed per-device overhead (CUDA context + per-device compute scratch) on every
participating GPU, beyond the slot-scaling compute buffer that is conserved across
the split. Measured ~0.9 GB/device on the Qwen3.6-27B GGUF (b9625), independent of
--parallel: layer-split TOTAL VRAM grew +894 MiB (parallel=8) / +946 MiB
(parallel=1) per extra GPU, ~linear to +2.6 GB at 4 GPUs. The fit folded a single
compute buffer for all subset sizes, so a k-GPU layer split was short by
~(k-1)*0.9 GB and could pin a context that fits the pool on paper but OOMs a device.
Reserve (k-1) * _PIPELINE_PER_DEVICE_OVERHEAD_MIB per subset in the layer-split fit;
k=1 adds nothing, so single-GPU sizing (and the validated benchmark rows) is unchanged.
B. Track the effective --spec-type. _extra_args_requests_mtp returned true on the
first MTP-ish --spec-type and consulted LLAMA_ARG_SPEC_TYPE even when a CLI
--spec-type was present, contrary to llama.cpp (last CLI value wins; a CLI flag
overrides the env). So `--spec-type draft-mtp --spec-type ngram-mod` or a non-MTP
CLI value with a stale MTP env over-reserved a drafter the launch won't load
(shrinking context / selecting extra GPUs). Route both detectors through a new
_effective_spec_type helper.
C. Keep known drafter weights in the fallback reserve. When a separate drafter's KV
metadata can't be sized, _estimate_mtp_overhead_bytes returned None and discarded
the drafter's known weight bytes, falling back to the flat 5% reserve; a drafter
larger than that cushion could launch over budget and OOM. Reserve the known
weights even when KV sizing fails (None only when nothing is known).
D. Restore quantized KV on tensor->layer-split downgrade. The tensor attempt drops a
quantized KV cache (tensor mode aborts on it). When the GPU-count or capacity gate
then downgrades to layer split -- which supports quantized KV -- the dropped type
was lost and the launch used f16, using more VRAM and shrinking context. Remember
the dropped type and restore it on downgrade (the launch re-emits it from the var).
Adds regression tests for each.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: per-device overhead in GPU pin, skip CPU draft, gate env spec-type
Address PR review feedback (three follow-up findings):
F1. Reserve the per-device layer-split overhead in the pin path too. The earlier
per-device reserve was added to the auto-context fit loops but not to
_select_gpus, which the explicit-ctx and file-size-only paths use to PIN GPUs
with -ngl -1 (no --fit fallback). A 2+ GPU pin within ~1 GiB/extra-GPU of the
budget could OOM a device at load. Add a per_device_overhead_bytes arg to
_select_gpus so a k-GPU pin must hold model + (k-1)*overhead; pass the pipeline
overhead at both pin call sites. Single-GPU pins are unchanged.
F2. Don't charge a CPU-offloaded drafter against the GPU budget. A user passing
--spec-draft-ngl 0 or --spec-draft-device none/cpu keeps the separate draft
model's weights + KV on CPU, but the budget still charged the full drafter GGUF
size, auto-reducing context or downgrading GPU selection. Detect the CPU-offload
flags and drop the separate drafter (and its flat fallback) from the budget; an
embedded head follows the main -ngl and is unaffected.
F3. Consult LLAMA_ARG_SPEC_TYPE only when it can reach the child. llama-server's CLI
args override env, and _build_speculative_flags emits a --spec-type/--spec-default
for every UI mode except "off". So a stale MTP env on a non-MTP model (auto mode)
made the fit reserve MTP that the emitted --spec-default disables, shrinking
context / picking extra GPUs. Gate the env consult on "no user --spec-type and UI
mode off"; the MTP-model auto path still engages via Studio's own detection.
Adds regression tests for each.
* studio: drafter budget precedence and --spec-default in effective spec-type
Two spec-precedence fixes surfaced by an independent multi-reviewer pass:
R3. Size the drafter the launch actually loads. _mtp_draft_for_budget consulted
LLAMA_ARG_SPEC_DRAFT_MODEL (via _extra_args_mtp_draft_path's env fallback)
before Studio's resolved mtp_draft_path, but _build_speculative_flags emits
--model-draft mtp_draft_path, which overrides the env at launch. With a stale
(smaller) env drafter, the budget under-reserved and could OOM. Order the
budget by what actually launches: CLI extras --model-draft (appended last,
wins), then Studio's emitted mtp_draft_path (when MTP engages and the user
doesn't own --spec-type), then the env drafter.
R4. Treat --spec-default as a CLI spec override in _effective_spec_type. It only
recognized --spec-type, so extras=["--spec-default"] with LLAMA_ARG_SPEC_TYPE=
draft-mtp fell through to the env and over-reserved MTP, even though the CLI
--spec-default overrides the env to a non-MTP default. Recognize it as a CLI
spec flag (resolves to "default", non-MTP) that suppresses the env fallback.
Adds regression tests for each.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: refine MTP draft reserve (parallel slots, last-wins, KV cushion, ranking)
Address PR review feedback (five follow-up findings, all edges of this session's
earlier MTP/auto-fit changes):
G1. Price the separate drafter's KV per --parallel slot. _mtp_draft_kv_bytes called
the drafter's _estimate_kv_cache_bytes with the default n_parallel=1, but the
drafter is served under the main model's slot count; a sliding-window drafter
(Gemma) grows KV per slot and was under-reserved. Thread n_parallel through the
draft KV / overhead estimate and the fit closure.
G2. Honor last-wins for the draft-offload flags. _extra_args_draft_offloaded_to_cpu
returned True on the first CPU value, so --spec-draft-ngl 0 --spec-draft-ngl -1
(final = GPU) wrongly dropped the drafter reserve while the server kept it on
GPU -> OOM. Decide on the final value of each flag only.
G3. Keep the flat cushion when only the drafter weights could be sized. The weights
fallback installs mtp_overhead_fn, which made callers drop the flat MTP reserve,
leaving the still-unsized draft KV with no cushion. Keep the flat fraction on in
that weights-only case, on top of the byte-accurate weights.
G4. Rank auto/cap GPU subsets by the active budget fraction. The ranking used a
hard-coded 0.95 while the fit tests _pin_fraction (lowered by the flat MTP
reserve); on mixed-total GPUs that could order subsets differently and pick a
worse plan. Rank with the same fraction the fit uses.
G5. Keep the embedded-head flat reserve under a draft CPU-offload flag. F2's
not-_draft_on_cpu guard also dropped the reserve for an embedded MTP head, which
is part of the main model and stays on GPU regardless of --spec-draft-ngl. Only
suppress the flat reserve for a CPU-offloaded separate drafter (no embedded head).
Adds regression tests for each.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: keep GPU on non-integer total, keep tensor flat reserve for weights-only
Two review findings:
- _get_gpu_memory dropped a whole GPU when nvidia-smi reported a non-integer
memory.total ("N/A" on some drivers / MIG / vGPU): index, free and total were
parsed in one try/except that skipped the line on any ValueError, so the GPU
vanished from the probe and the load could silently spill to CPU. Parse index
and free (required) first, then total separately, defaulting to 0 (the fit then
uses the free*frac path for that GPU). Adds N/A and bad-free test cases.
- Tensor planning skipped the flat MTP reserve for a weights-only drafter (file
size known, KV unsizable): the capacity gate used the byte floor whenever
mtp_overhead_fn was set, so it reserved only the drafter weights and no draft
KV. Tensor mode has no --fit valve, so that could overcommit and OOM. Keep the
flat reserve (never below the byte floor) in the weights-only case too, mirroring
the layer-split _mtp_kv_unsized handling. Adds a regression test.
(A third suggestion -- fold --batch-size into the compute-buffer reserve -- was
checked on hardware and declined: -b 8192 -ub 512 used identical VRAM to the
default at -c 64000, so the logical batch does not size the graph buffer; the
estimate correctly uses the physical micro-batch.)
* studio: budget the main KV from LLAMA_ARG_CACHE_TYPE env when Studio emits none
The child inherits LLAMA_ARG_CACHE_TYPE_K / LLAMA_ARG_CACHE_TYPE_V, but Studio
emits --cache-type-k/-v only when the param or extras set the type. When neither
does, a heavier env type (f32) reaches the child while the auto-fit budget
assumed the f16 default, under-reserving the main KV and risking OOM at the
advertised context. This is the one main-KV axis that lacked the env-aware
handling the other axes already have (spec-type, draft model, draft cache type,
ubatch).
load_model now adopts the heavier of the two env types when it exceeds f16 (only
f32 does), and the launch re-emits it so child and budget stay byte-consistent.
Quantized env types are <= f16 and remain safely over-reserved by the default,
so they are left untouched (no change). A single value is used because the
budget's KV estimate has one cache_type_kv knob, matching parse_cache_override's
existing key/value collapse.
Adds _env_main_cache_type_for_budget plus regression tests covering f32 adoption,
the K/V heavier-of collapse, quantized/unknown no-ops, and the load_model source
precedence.
* studio: budget tensor parallel when LLAMA_ARG_SPLIT_MODE env selects it
Studio emits --split-mode tensor only on its tensor branch; the default
layer-split path emits nothing and resolve_tensor_parallel consults only extras.
The child inherits LLAMA_ARG_SPLIT_MODE, so a tensor env on a layer-split plan
silently runs the child tensor-parallel (heavier per-device compute buffer)
while the budget reserved only the layer-split per-device overhead, under-
reserving on multi-GPU.
load_model now flips the plan to tensor when extras do not set a split mode and
the env selects tensor, so Studio plans, reserves, and emits tensor consistently.
The flip is one-directional (guarded on not tensor_parallel and no extras
split-mode) so an existing tensor plan is never downgraded and extras keep
precedence. Other env modes (layer/row/none) are not a runtime-heavier surprise
and are left untouched.
Adds _env_split_mode_is_tensor plus unit and load_model source-level tests.
* studio: reconcile inherited llama.cpp env with the budgeted launch decision
Addresses a review pass over the VRAM auto-fit work. The budget now sizes the
right amount, but the child process inherits LLAMA_ARG_* env (see
child_env_without_native_path_secret), and a few axes could still run the child
in a mode Studio neither chose nor budgeted.
Mixed known/unknown GPU totals over-advertised the pooled layer-split budget.
_pool_budget_mib pooled free and total separately, so an unknown-total GPU
(MIG/vGPU/N/A) contributed its full free with no cushion when mixed with
known-total GPUs (~(1-frac)*free over-advertise, about 500 MiB in a two-GPU
case). It now sums each GPU's own usable budget, and the layer-split fit calls
take that as an absolute budget (budget_frac=1.0, total_mib=None) so the fit and
the footprint check agree. All-known-total pools are unchanged.
LLAMA_ARG_SPLIT_MODE=tensor survived a tensor-to-layer downgrade. The downgrade
only stripped CLI extras, so the inherited env still ran the child tensor while
Studio budgeted layer split. When the final decision is layer split, a non-layer
inherited split mode (and any paired LLAMA_ARG_TENSOR_SPLIT) is now cleared from
the child env.
Inherited quantized LLAMA_ARG_CACHE_TYPE_K/_V crashed tensor mode. Tensor mode
aborts on a quantized KV cache; Studio drops a quantized cache_type_kv for the
tensor attempt but the inherited env reached the child anyway. When the final
decision is tensor split, a quantized cache-type env is now cleared so the child
uses the tensor-safe default that was budgeted.
Env-derived cache budget no longer mutates the emitted launch flags. An env-only
main KV type now informs the budget only; it is not re-emitted, so an asymmetric
K=f32,V=f16 env reaches the child as set instead of being rewritten to symmetric
--cache-type-k/-v f32.
Adds source-level regression tests for all four and confirms the documented
single-GPU/tensor/pipeline numbers are byte-identical before and after.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: tighten comments in the VRAM auto-fit code
Compress the verbose docstrings and inline comments added by this work to
succinct 2-4 line versions, drop restated/obvious ones, and cut duplicated
rationale across the two tensor-downgrade branches. Keeps the non-obvious intent
(env-inheritance precedence, the #24102 embedded-draft floor, the per-device
overhead and pool-budget rationale) while removing roughly 120 lines of comment
text from llama_cpp.py. Also trims the few longest test-comment blocks; concise
per-test scenario notes are left intact.
No logic change: verified with comment_tools.py check --strip-docstrings (code-
only signature unchanged vs the prior commit) and the full backend suite still
passes (824).
* studio: lock in env-drafter engagement for the separate-draft reserve
A review suggested an env-provided LLAMA_ARG_SPEC_DRAFT_MODEL would skip the
draft reserve and OOM. It does not: the gate's _extra_args_mtp_draft_path(extra_args)
call defaults env=None, which consults os.environ, so an env-only drafter still
sets _user_draft_via_extras and is sized via _env_draft_for_budget. Add a source
guard that the gate keeps the env-inclusive form (not extras-only env={}) and a
behavioral test mirroring the reviewed scenario, so a future cleanup can't
regress it. No production change.
* studio: carry the unsized MTP reserve and env split/offload into tensor planning
Addresses a review pass over the multi-GPU and env-inheritance paths.
Tensor planner dropped the unsized draft-KV cushion. When a separate drafter has
known weights but unreadable KV metadata, _plan_tensor_parallel receives a
non-None weights-only mtp_overhead_fn and applied the flat 2 GiB reserve only for
the no-fn case, so its binary search spent the unsized-KV cushion on context and
over-advertised. Add mtp_flat_reserve_bytes (subtracted from the pooled budget and
the even-split check), and pass it from load_model whenever _mtp_kv_unsized. The
layer path and the tensor pre-gate already kept this cushion.
Stale LLAMA_ARG_TENSOR_SPLIT survived in tensor mode. When the planner picks an
even split it emits no --tensor-split, so an inherited tensor-split env reached the
child and overrode the budgeted split. The layer downgrade branch cleared it; the
tensor branch now does too.
Env-only draft CPU offload was ignored. _extra_args_draft_offloaded_to_cpu checked
extras but not LLAMA_ARG_N_GPU_LAYERS_DRAFT, so an env-offloaded drafter was still
charged GPU budget and under-advertised context. It now consults that env (the
device flag has no env), called with env=os.environ.
Layer-split compute buffer had no fallback when GGUF dims are missing. The estimate
returns 0 then, so the layer path folded no buffer while the tensor path falls back
to the flat reserve. Use the flat reserve for the layer path too (a safe upper
bound, since the tensor buffer >= the layer one).
All four are gated on conditions the documented benchmarks don't hit; the
single-GPU/tensor/pipeline reconfirm numbers are byte-identical, and the full
backend suite passes (830) with regression tests for each fix.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: share the env-aware tensor decision across load and dedup matchers
A review pass found the inherited-LLAMA_ARG_SPLIT_MODE=tensor flip lived only
in load_model, so the two duplicate-load matchers disagreed with it.
Consolidate the decision into _effective_tensor_parallel (extras + toggle, then
flip on when extras set no split mode and the child inherits a tensor split
env). load_model, the backend matcher (_already_in_target_state) and the route
matcher (_request_matches_loaded_settings) now all call it. Before, an env-driven
tensor server compared against resolve_tensor_parallel (env-blind) in both
matchers, so a follow-up load that should dedup was seen as a mismatch and the
healthy server was needlessly killed and reloaded.
Also finish the tensor cache-type handling: when the tensor attempt drops a
quantized KV it now re-adopts a heavier inherited env cache type (f32) for the
budget, mirroring the initial adoption; and the two layer-split downgrades clear
_cache_type_from_env so the restored quantized type is actually re-emitted rather
than left to a stale inherited env.
All gated on inherited env the documented benchmarks don't set; the single-GPU,
tensor and pipeline reconfirm numbers are byte-identical, and the full backend
suite passes (832) with unit + source regression tests for the shared helper and
the route matcher.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: complete the env-aware tensor/spec handling across all paths
A second review pass found the env-aware tensor/MTP handling was applied
asymmetrically: some paths inherited LLAMA_ARG_* env, others didn't. Three real
follow-ups, plus a small consolidation so the env semantics live in one place.
1. Tensor fallback ignored the inherited tensor env. load_with_tensor_fallback
computed its retry gate with the env-blind resolve_tensor_parallel, so an
env-only tensor load (toggle off, no --split-mode extra) that crashed on a
tensor-incompatible GGUF re-raised instead of retrying layer split. It now
uses the env-aware decision; and since the inherited env would otherwise
re-engage tensor on the retry (CLI args persist, the env does too), the retry
forces --split-mode layer (CLI wins over env) so it can't re-crash.
2. Duplicate-load matchers looped reloads after a tensor->layer downgrade. Both
matchers compared the env-expanded tensor decision against the loaded server,
but load_model may downgrade tensor to layer (capacity/buffer) and scrub the
child env. The still-set parent env then made every identical request look
like a mismatch, killing and reloading a healthy layer server. Add
_tensor_parallel_matches_loaded, which only lets an inherited tensor env raise
a match against a server that actually launched tensor; a downgraded server
matches the same request (an identical load would downgrade the same way).
3. MTP binary-capability fallback leaked an inherited LLAMA_ARG_SPEC_TYPE. When
the binary lacks MTP, _emit_mtp degraded but emitted no spec flag, so an
inherited LLAMA_ARG_SPEC_TYPE=draft-mtp still reached the child and attempted
MTP the gate had budgeted off. It now emits --spec-default (CLI wins over env)
like the sibling no-head / non-MTP fallbacks.
Consolidation: moved _env_split_mode_is_tensor / _effective_tensor_parallel into
llama_server_args.py (with the new _tensor_parallel_matches_loaded) so the
lightweight tensor_fallback module can share them without importing llama_cpp;
llama_cpp re-exports them for back-compat.
All gated on inherited env the documented benchmarks don't set; the single-GPU,
tensor and pipeline reconfirm numbers are byte-identical, and the full backend
suite passes (883) with regression tests for each fix.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: budget the heavier axis of asymmetric --cache-type-k/-v extras
A review pass found the explicit-extras counterpart of the env cache-type fix.
load_model adopts the heavier inherited LLAMA_ARG_CACHE_TYPE_K/_V env for the
reserve, but the explicit-extras path used resolve_cache_type_kv, which collapses
both axes to one last-wins value. So extras such as
--cache-type-k f32 --cache-type-v f16 (lighter axis last) budgeted f16 for both
axes while the child allocates f32 on K, over-advertising context and
re-opening the OOM path this PR closes.
Add parse_cache_override_per_axis (keeps the K/V last-wins values apart) and
_extra_args_main_cache_type_for_budget (the heavier of the two by bytes/elem),
and budget from it. The user's extras are appended last and win per axis at the
child, so this only raises the reserve; the emitted command and the asymmetric
child cache are unchanged, and the common single-axis / symmetric cases resolve
to the same type as before.
Reconfirm numbers (single-GPU table, tensor, pipeline) are byte-identical, and
the full backend suite passes (892) with per-axis parser and heavier-axis budget
regression tests.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: fix tensor-safety masking and strip inherited HF drafter selectors
A review pass found two more env/extras edge cases on the speculative and tensor
cache paths.
Tensor-safety could miss a quantized axis. The previous change budgets the
heavier-by-bytes cache type, but that masks a quantized axis paired with a
heavier one: --cache-type-k f16 --cache-type-v q4_0 resolves to f16, so the
tensor-safety block did not fire and the q4_0 axis survived into tensor mode,
which aborts on quantized KV. Test each explicit --cache-type-k/-v axis (not just
the budget type) so any quantized axis drops the cache for the tensor attempt.
Inherited HF drafter selectors were not stripped. _extra_args_mtp_draft_path
treats --spec-draft-hf / -hfd / -hfrd / --hf-repo-draft as drafter selectors, but
_SPEC_FLAGS only stripped the local --model-draft selectors, so on an inherited-
extras Apply a stale HF drafter survived and last-wins-overrode Studio's
re-derived spec choice. Add the HF aliases to _SPEC_FLAGS. The per-drafter tuning
knobs (--spec-draft-type-*, -ngld, --spec-draft-device) are intentionally left in
place: the VRAM budget reads them via the same parsers the child honors, so they
stay consistent on inherit, and stripping them would silently move a CPU-offloaded
drafter back onto the GPU.
A third flagged item -- that the HF draft env var should be LLAMA_ARG_HFD_REPO --
was a false positive from a stale manpage; the bundled binary's common/arg.cpp
sets LLAMA_ARG_SPEC_DRAFT_HF_REPO for --spec-draft-hf, which the code already
uses, so it is left unchanged.
Reconfirm numbers (single-GPU table, tensor, pipeline) are byte-identical, and
the full backend suite passes (899) with regression tests for both fixes.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* studio: preserve asymmetric cache on tensor downgrade and skip CPU-drafter reserve
A review pass found two more tensor-path edges, one a regression from the
per-axis cache change.
Tensor-to-layer downgrade collapsed asymmetric cache extras. The per-axis
tensor-safety check strips an asymmetric --cache-type-k/-v (tensor rejects
quantized KV), but the downgrade restored only the scalar heavier type, so a
layer fallback silently rewrote --cache-type-k q4_0 --cache-type-v f16 to
symmetric f16/f16 even though layer split supports the original. Save the
original extras before the tensor strip and restore them verbatim (minus the
user --split-mode) on both downgrade points; the budget still uses the heavier
scalar, the child gets the real asymmetric cache. Before the per-axis change this
case happened to survive (last-wins was f16, untouched), so this restores that.
Tensor mode reserved GPU VRAM for a CPU-offloaded drafter. The layer path drops
the flat MTP reserve when the only drafter is a separate CPU one with no embedded
head, but the tensor capacity gate and planner still charged it, under-advertising
context. Gate the tensor reserve on the same condition via _mtp_reserves_gpu.
Reconfirm numbers (single-GPU table, tensor, pipeline) are byte-identical (both
fixes are gated on conditions the benchmarks don't hit), and the full backend
suite passes (901) with regression tests for each.
* studio: drop now-unused llama_server_args imports from llama_cpp
The refactor re-pointed load_model and the matchers off resolve_tensor_parallel /
resolve_cache_type_kv and moved the env split-mode helper into llama_server_args,
leaving those three names imported but unused in llama_cpp. The repo's import-hoist
safety-net lint blocks that, so drop them; the env split-mode test now imports
_env_split_mode_is_tensor from its real home (llama_server_args).
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Studio: Add Tensor-Parallel llama.cpp support
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: harden Tensor-Parallel fallback and GPU selection
* Studio: reconcile split-mode extras and harden tensor-split planning
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: reconcile split-mode extras in backend duplicate-load guard
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: preserve inherited non-tensor split modes on reload
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: honor cancellation in tensor fallback, preserve tensor mode on rollback, and don't raise an explicit small context
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Studio: reconcile split-mode in reload check and strip it on tensor downgrade
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Strip --tensor-split alongside --split-mode so inherited ratios don't override the tensor planner
An inherited or stale --tensor-split in llama_extra_args was appended after
Studio's computed --tensor-split and won last in llama.cpp, re-introducing the
asymmetric-GPU OOM tensor mode is meant to prevent. Group -ts/--tensor-split
into the split-mode shadow set so it is stripped on inherit and on the layer
fallback; parse_split_mode_override still keys on the mode value only.
* Drop quantized KV for the tensor attempt and report native max context
Tensor mode aborts on a quantized KV cache, so a user with q8_0/q4_1 etc. who
enabled Tensor Parallelism silently fell back to layer split. Clear the cache
type (and strip inherited/explicit --cache-type) for the tensor attempt only;
the layer fallback re-runs with tensor off and keeps the user's choice.
Also report max_available_ctx from the native context, not an explicit small
-c, so the context slider no longer warns too early in tensor mode.
* Reconcile inherited split-mode extras in the already-loaded check
When a same-model load omitted llama_extra_args, the tensor comparison resolved
the raw (None) request and treated an inherited --split-mode tensor server as a
mismatch, forcing a needless reload. Compare using the stored extras stripped
the same way the reload strips them.
* Pass tensor_parallel through compare-mode loads
The generalized compare path loaded each GGUF without tensor_parallel, so
compare ran layer split even with the toggle on and left the settings sheet
stale. Send the toggle and hydrate the loaded state from the response, matching
the main chat and recipe load paths.
* Add --tensor-parallel flag to unsloth studio run
The headless one-liner could only reach tensor mode by passing --split-mode
tensor as a raw llama.cpp extra. Add a first-class --tensor-parallel/
--no-tensor-parallel option that sets the tensor_parallel field on the
/api/inference/load payload, forwarded through the studio-venv re-exec like the
other polarity flags. Matches the web UI toggle and the API field.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: danielhanchen <michaelhan2050@gmail.com>
oobabooga